-
Notifications
You must be signed in to change notification settings - Fork 304
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Version v4.13.3 is making HA fail #2579
Comments
yes this updates make my system ERROR dont install this update!!! |
Having the same issue! Had to reboot in safe mode (which took a few trials) and deactivate the integration. It seems like then installing the old version makes HA stable again. |
OUT OF MEMORY error and HA reboots every 5 minutes. |
after last update 2024-10-03 08:58:18.551 WARNING (MainThread) [homeassistant.util.loop] Detected blocking call to load_verify_locations with args (<ssl.SSLContext object at 0x7effb6ae7b50>, '/usr/local/lib/python3.12/site-packages/certifi/cacert.pem', None, None) inside the event loop by custom integration 'alexa_media' at custom_components/alexa_media/init.py, line 341: AlexaLogin( (offender: /usr/local/lib/python3.12/ssl.py, line 708: context.load_verify_locations(cafile, capath, cadata)), please create a bug report at https://github.com/alandtse/alexa_media_player/issues |
Same problem here HA restarts every 5min or when you give any commands to HA. Rollback to 4.13.2 solves the issue. The issue is NOT related to 2024.10.0 I am still on 2024.09.03 and have the problem. I highly recommend to completely restart (including OS) after the rollback to version 4.13.2. |
I wondered why my HA was rebooting every couple of minutes :D Rebooted the host and managed to get HACS to download 4.13.2 before it had chance to load the integration. |
Same here, HA kept rebooting. Managed to disable the integration, now running stable. I have not downgraded the integration yet, see how long a fix takes. |
Same Issue - had to restore HA - will skip that version for now ;) |
Same issue for me. Just rolled back to the previous version. Let’s see how stable it remains after. |
+1. 4.13.3 kills my HA also. Rolled back to 4.13.2 |
Same issue here. Its so bad I can't even SSH into my server now to kill home assistant :( |
HA seems to hang as soon as this integration receives an event from Alexa. With debug logging enabled one of the last log messages is:
I'm able to reproduce this either by opening the Alexa app or by saying "Alexa, what's the time?" |
Same issue! HA 2024.9.3 with v4.13.3 is killing my HA setup. CPU and memory usage explodes and HA reboots every few minutes! |
thanks for that information. Alexa MediaPlayer is now deactivated... |
@ChristophCaina The update just came out a few hours ago. One solution for now is to "deactivate" Alexa Media Player, but other solution would be to roll back to the previous version of AMP which is still working fine. |
Had the same issue with
I got this from the supervisor, but not sure if it is related.
|
Can confirm, do not pull the trigger on this update : o Core 2024.10.0 |
Silly newbie question: how did you roll back, please? Did you do it from HACS and simply select the earlier version? I get a warning if I do that, but I assume that I ignore that? |
also successfully rolled back to 14.3.2 - did the trick perfectly! PS: This is the issue in ha-core since many guys thought the problem was related to the latest 2024.10.0 Update: |
Spent last hour looking why my HA keep crashing every 20 minutes after updated it to 2024.10... |
@npa-homeassistant The tricky part can be to rollback before your HA stops responding. Seems it starts using up all the CPU and memory. I would guess those running on a raspberry pi would have less time to do the roll back than those that have a more robust system. Being that you are a newbie like myself, my suggestion would be to do backups often. You don't need 20 different backups. I usually always have two backups. If I do a new backup today, i keep the one I had done before, but delete the rest. |
This is b
I suggest to just immediately disable the alexa-media-player integration so you've time to roll back to the older version Cheers mate |
Same here. CPU temperature went over 70 degrees Celsius. |
This is definitely true, I can confirm this. Running HA as a virtual machine on a Synology NAS I had to ramp up the VMs memory to be fast enough to react inside HA. |
@florianmulatz I am a newbie, and I keep learning from comments like yours. My 2 cents to your last comment is that once you do roll back to the correct version, the Alexa Meida player integration will still be disabled, and it won't show up on the regular devices and services page. Well it does show up, but it would be hidden on the right side where you have to click to show disabled integrations. I had disabled AMP for other reasons earlier on the day, and I couldn't seem to figure where to find it. On the top right it shows "1 disabled show", for newbies like me it is easy to miss |
on the button right top on you screenshot "SHOW" |
Hi... Solarflow...Zendure😆😆😆 |
14.3.3 appears to have been yanked to avoid anymore problems with it. Little excitement in our others smooth smart home life. |
Thanks Team. Disabling Alexa worked for me. |
Ended up having to delete this integration to get HASS to boot. Hopefully this will be fixed soon. |
Not likely is it? |
There's already a PR which fixes it. One line code change: https://github.com/alandtse/alexa_media_player/pull/2589/files Would be helpful if others would try it, and confirm this resolves it for them too. |
+1 |
@lviciedo HOW DARE YOU! |
same here. HA reload constant and out of memory:
|
After reading this I just wanted to stop by and say a BIG THANK YOU to @danielbrunt57! I appreciate every single second of the time you've put in working on this integration. You are an absolute gentlemen sir! This integration is wonderful and a much enjoyed part of my Home Assistant setup. |
If you've come here to say "ME TOO":There is a proposed fix, it is a simple change to one file. This integration is maintained by volunteers, so please help them by trying the fix rather than just disabling the integration. I can tell you it has worked for me. https://github.com/alandtse/alexa_media_player/pull/2589/files It appears to be specific to certain devices, so a community effort here would get the fix verified more quickly. The code maintainers obviously don't have the affected devices. |
You are absolutely right, in every single word. I really appreciate your work and dedication. Please feel encouraged to further keep this integration alive. Don‘t give a cent on comments of this ungrateful fools which don‘t understand volunteership. It also took me some time to figure out which integration messes up, but thats not your fault but rather mine or the not to easy to read logs from homeassistant. |
I do not see how that PR would fix the infinite loop. Adding [:] to a list keeps the list identical and that's the only change in the PR. e.g.
is identical to
I think we have too many people relying on AI to submit PRs on this. Although it is a way to keep the list static during the for loop. Looking at the comments it may actually fix it. However the original PR is still misusing the HA API. |
It's not the same. The [:] means it's going to be iterating over a copy of the list and that's the critical difference. This is not an AI fix. I found it myself by analysis and testing. I do however share your wider concern. If you mean the original code is misusing HA by doing blocking calls then you're right, but that's a separate issue, or have I misunderstood what you meant? |
I, too, found my HA crashing repeatedly due to the bug. But I’m not here complaining at the devs who put in their hard work for free. Especially for a HACS integration, which explicitly comes with a level of potential instability or risk that you must accept when installing it. I installed the latest version without waiting, without seeing how other people fared with it, and predictably that made me a Guinea pig. If you don’t want to be a Guinea pig, if you don’t want to be subject to these sorts of bugs, don’t install updates right away. In fact, maybe HACS just isn’t for you. Don’t come here bitching at the devs that put their metaphorical blood, sweat, and tears into free software for you. |
Yes I meant the original PR. And yes I amended my comment about it being the same. |
same here reboot automatically every minutes .... |
https://github.com/alandtse/alexa_media_player/pull/2589/files |
Did the one line change in the helper.py file. It’s working without any issues since last 4 hours. No crashing any more.On Oct 5, 2024, at 2:50 PM, Petro31 ***@***.***> wrote:
Not likely is it?
There's already a PR which fixes it.
One line code change:
https://github.com/alandtse/alexa_media_player/pull/2589/files
Would be helpful if others would try it, and confirm this resolves it for them too.
I do not see how that PR would fix the infinite loop. Adding [:] to a list keeps the list identical and that's the only change in the PR.
e.g.
items = ['a','b']
for item in items:
...
is identical to
items = ['a','b']
for item in items[:]:
...
I think we have too many people relying on AI to submit PRs on this.
It's not the same. The [:] means it's going to be iterating over a copy of the list and that's the critical difference.
This is not an AI fix. I found it myself by analysis and testing. I do however share your wider concern.
If you mean the original code is misusing HA by doing blocking calls then you're right, but that's a separate issue, or have I misunderstood what you meant?
Yes I meant the original PR. And yes I amended my comment about it being the same.
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you commented.Message ID: ***@***.***>
|
I had some mixed results earlier today, but I just rebooted again with the one line change and all seems well. I also added a log within the loop to output the |
Some feedback on this : I have HA 2024.9.3 with v4.13.3. Same problem, one reboot every 15 minutes. I applied the one line fix #2589. Seems stable again, no reboot since. |
I am happy to give it a test (I have 13 different alexa devices) but have already rolled back so don't have the updated version available to make the change. I am not a coder but have studio code server installed. Is there a package I could download that includes the fixed file for testing? |
Same for me - seems to be stable for at least an hour |
@snigehere All custom integrations that you have downloaded are available as they are in their GitHub repository on your HA under config/custom_components. HACS is only a download vehicle. |
Same here, this fix seems to correctly solve the problem |
IMPORTANT: Please search the issues, including closed issues, and the FAQ before opening a new issue. The template is mandatory; failure to use it will result in issue closure.
Describe the bug
I installed v4.13.3 and shortly after my HA kept loosing connection. I would re boot, it would work fine for a few minutes, then it would disconect.
To Reproduce
I used a full backup for HA for Sept 30th. Before the Alexa update I did a core update, assumed that caused the issue. Once I restored I didn't do the core update but did do the Alexa update. Few minutes later issue of disconnect started again.
Restored the Sept 30th backup again. This time I did the HA core update, but skipped the v4.13.3 update for Alexa. My HA has been stable for over 30 minutes now.
Expected behavior
I would expect to install v4.13.3 and not have my HA system crash.
Screenshots
No screen shots
System details
const.py
or HA startup log): Stable version 4.13.2pip show alexapy
in homeasssistant container or HA startup log):Debug Logs (alexa_media & alexapy)
Please provide logs.
Additional context
The text was updated successfully, but these errors were encountered: